The selective delivery of drugs has long been a goal in the treatment of cancer. Antibody-mediated delivery of conjugated drugs, radionuclides, and toxins have been investigated in the recent past, and in all cases it is clear that improvements are required to widen the therapeutic window and to reduce toxicity. Each of the above conjugates has advantages and disadvantages, but we have chosen to focus our attention on the use of toxins (recombinant ricin A chain, rRA) as the inhibitory agent within the conjugate. Such immunotoxins are catalytic in contrast to the single hit characteristics of the other two conjugates. Imaging of breast cancer antibodies in animals and in human patients with breast cancer shows that the antibodies reach their targets, and this has been confirmed with biodistribution studies using immunotoxins (conjugates of monoclonal antibodies and toxins) in animals. Immunotoxins limit or prevent the growth of tumors in animal models, but the doses that are required are often too near toxic doses. Immunotoxins will undoubtedly be used in combination with other therapeutic drugs, especially elimination of pockets of metastatic disease after debulking therapy. Current results in Phase 1/11 trials indicate that more immunotoxin needs to be given to patients in order to have better therapeutic effects; therefore, toxicity must be reduced. We believe that the approach proposed using engineered antibodies will change the pharmacokinetic and biodistribution properties of immunotoxins and will increase the efficacy of immunotoxins. Our methods go beyond the limitations of natural monoclonal antibodies and develop procedures for refining the immunotoxin by genetically manipulating its parts. This RFA enables a greatly increased emphasis on the development of much more comprehensive approach than would have been possible without that support. The scale of the work will increase the chances that a drug still in the early stages of development can be improved significantly. It permits scientific issues, such as the structure/function relationships investigated in Program 3, to be explored in a more thorough and rational manner instead of making one conjugate in the hopes that it will work. Reduction in the size of the antibody portion of the immunotoxin and elimination of the carbohydrate on both the antibody and the toxin (recombinant ricin A chain) are likely to eliminate many of the problems associated with immunotoxins. Our target is breast cancer, and because the antibody we have chosen cross reacts with ovarian cancer, we can also potentially treat ovarian cancer. By cloning lgG antibodies specific for breast cancer, the constant regions of lgG can be eliminated and the variable regions, which form the antibody combining site, can be converted into single chain Fv fragments or biosynthetic antibody binding sites (BABS). BABS is the smallest possible antibody fragment, can be made into a divalent form of BABS, has the potential for manipulation of the antibody combining site (affinity), and is produced in E. coli. By coupling the antibody chemically (XLIT) or genetically (gene fused immunotoxin, GFIT to recombinant ricin A chain, immunotoxins will be studied for their biophysical properties. Our approach will create a systematic series of immunotoxins containing mono-, di- and multivalent BABS, Fab-SH, F(ab')2, and intact lgG that will be compared and tested thoroughly for their effects on breast and ovarian cancer and control cells in vitro and in vivo.